Controlling submission of content

1. An apparatus comprising:

at least one processor; and

at least one memory including computer program code,

the at least one memory and the computer program code configured to, with the at least one processor, cause the apparatus at least to perform:

obtaining a value of one or more parameters which vary with actions of a user that are indicative of a mood of the user;

accessing an artificial intelligent agent configured to use one or more trained machine learning models and selecting the one or more trained machine learning models, from a plurality of differently trained machine learning models, based on the obtained value of one or more parameters which vary with actions of the user that are indicative of the mood of the user, the plurality of differently trained machine learning models being configured to provide respective outputs;

providing content composed by the user as an input to the artificial intelligent agent to cause: (i) a selection between a first process for controlling a submission based on the content composed by the user and a second, different process for controlling the submission based on the content composed by the user, and (ii) generation of feedback to the user on the content composed by the user, the feedback being dependent on the artificial intelligent agent, wherein the selection between the first process and the second process is dependent on the artificial intelligent agent;

controlling the submission based on the content composed by the user; and

causing the feedback to be provided to the user,

wherein the second process provides the feedback before enabling the submission based on the content composed by the user,

wherein the first process for controlling the submission based on the content composed by the user provides for automatic submission based on the content composed by the user, and

wherein selecting the one or more trained machine learning models comprises selecting the one or more trained machine learning models that have been trained on previous content composed by the user in a different mood profile than a present mood profile of the user.

2. The apparatus as claimed in claim 1 wherein the at least one memory and the computer program code are configured to, with the at least one processor, cause the apparatus at least to further perform: provide another input, based on a subject that is addressed by the content composed by the user, to the artificial intelligent agent,

wherein the selected one or more trained machine learning models are configured to generate content addressing the subject, wherein the content is generated to simulate content composed by the user or to simulate a response by the user to an input, and

wherein the selection between the first process and the second process for controlling the submission based on the content composed by the user is dependent on a comparison, by the artificial intelligent agent, of the generated content with the content composed by the user.

3. The apparatus as claimed in claim 2 wherein the subject is obtained from further content which has been provided to the user, wherein the further content has been rendered to the user in a period of time or period of user interaction time directly preceding composition of the content by the user.

4. The apparatus as claimed in claim 2 wherein the subject is obtained by processing the content composed by the user.

5. The apparatus as claimed in claim 1 wherein the selected one or more trained machine learning models are configured to generate content addressing the content composed by the user, and

wherein the selection between the first process and the second process for controlling the submission based on the content composed by the user is dependent on a sentiment classification, by the artificial intelligent agent, of the generated content.

6. The apparatus as claimed in claim 1 wherein the at least one memory and the computer program code are configured to, with the at least one processor, cause the apparatus at least to further perform: apply the selected one or more trained machine learning models to the content composed by the user.

7. The apparatus as claimed in claim 1 wherein the plurality of differently trained machine learning models are trained using training data comprising previous content composed by the user.

8. The apparatus as claimed in claim 7 wherein the at least one memory and the computer program code are configured to, with the at least one processor, cause the apparatus at least to further perform:

classify the previous content composed by the user according to a value, obtained when the previous content was composed, of the one or more parameters which vary with actions of the user;

form different sets of training data from differently classified previous content; and

train respective different machine learning models on respective different sets of training data so as to provide the plurality of differently trained machine learning models.

9. The apparatus as claimed in claim 7 wherein the one or more parameters which vary with actions of the user comprise parameters which parameterize mood profiles of the user.

10. The apparatus as claimed in claim 7 wherein the one or more parameters which vary with actions of the user comprise one or more intended recipients of the content composed by the user, and

wherein the selection of one or more trained machine learning models comprises selecting one or more trained machine learning models which have been trained on the previous content composed by the user for a similar one or more intended recipients.

11. The apparatus as claimed in claim 1 wherein providing the content composed by the user as the input to the artificial intelligent agent is responsive to a user input requesting a submission based on the content composed by the user.

12. The apparatus as claimed in claim 1 wherein the one or more trained machine learning models are comprised in one layer of the artificial intelligent agent, and wherein one or more other layers comprise one or more evaluative modules configured to evaluate one or more respective outputs of the selected one or more trained machine learning models.

13. The apparatus as claimed in claim 1 wherein the selected one or more trained machine learning models are trained to generate content as if composed by the user.

14. The apparatus as claimed in claim 1 wherein the submission is disabled until the content composed by the user has been changed by the user.

15. The apparatus as claimed in claim 14 wherein the submission is disabled until the feedback changes in response to the change to the content composed by the user.

16. The apparatus as claimed in claim 1 wherein the submission is disabled for a period of time following the provision of the feedback to the user.

17. A non-transitory computer readable medium comprising program instructions stored thereon for performing at least the following:

obtaining a value of one or more parameters which vary with actions of a user that are indicative of a mood of the user;

accessing an artificial intelligent agent configured to use one or more trained machine learning models and selecting the one or more trained machine learning models, from a plurality of differently trained machine learning models, based on the obtained value of one or more parameters which vary with actions of the user that are indicative of the mood of the user, the plurality of differently trained machine learning models being configured to provide respective outputs;

providing content composed by the user as an input to the artificial intelligent agent to cause: (i) a selection between a first process for controlling a submission based on the content composed by the user and a second, different process for controlling the submission based on the content composed by the user, and (ii) generation of feedback to the user on the content composed by the user, the feedback being dependent on the artificial intelligent agent, wherein the selection between the first process and the second process is dependent on the artificial intelligent agent;

controlling the submission based on the content composed by the user; and

causing the feedback to be provided to the user,

wherein the second process provides the feedback before enabling the submission based on the content composed by the user,

wherein the first process for controlling the submission based on the content composed by the user provides for automatic submission based on the content composed by the user, and

wherein selecting the one or more trained machine learning models comprises selecting the one or more trained machine learning models that have been trained on previous content composed by the user in a different mood profile than a present mood profile of the user.

18. The non-transitory computer readable medium as claimed in claim 17 wherein the program instructions are further configured to provide another input, based on a subject that is addressed by the content composed by the user, to the artificial intelligent agent,

wherein the selected one or more trained machine learning models are configured to generate content addressing the subject, wherein the content is generated to simulate content composed by the user or to simulate a response by the user to an input, and

wherein the selection between the first process and the second process for controlling the submission based on the content composed by the user is dependent on a comparison, by the artificial intelligent agent, of the generated content with the content composed by the user.

19. The non-transitory computer readable medium as claimed in claim 17 wherein the selected one or more trained machine learning models are configured to generate content addressing the content composed by the user, and

wherein the selection between the first process and the second process for controlling the submission based on the content composed by the user is dependent on a sentiment classification, by the artificial intelligent agent, of the generated content.

20. A system comprising:

a plurality of differently trained machine learning models, the plurality of differently trained machine learning models being configured to provide respective outputs;

an artificial intelligent agent;

at least one processor; and

at least one memory including computer program code,

the at least one memory and the computer program code configured to, with the at least one processor, cause the system at least to perform:

obtaining a value of one or more parameters which vary with actions of a user that are indicative of a mood of the user;

selecting one or more trained machine learning models, from the plurality of differently trained machine learning models, based on the obtained value of one or more parameters which vary with actions of the user that are indicative of the mood of the user, wherein the artificial intelligent agent is configured to use the selected one or more trained machine learning models;

providing content composed by the user as an input to the artificial intelligent agent to cause: (i) a selection between a first process for controlling a submission based on the content composed by the user and a second, different process for controlling the submission based on the content composed by the user, and (ii) generation of feedback to the user on the content composed by the user, the feedback being dependent on the artificial intelligent agent, wherein the selection between the first process and the second process is dependent on the artificial intelligent agent;

controlling the submission based on the content composed by the user; and

causing the feedback to be provided to the user,

wherein the second process provides the feedback before enabling the submission based on the content composed by the user,

wherein the first process for controlling the submission based on the content composed by the user provides for automatic submission based on the content composed by the user, and

wherein selecting the one or more trained machine learning models comprises selecting the one or more trained machine learning models that have been trained on previous content composed by the user in a different mood profile than a present mood profile of the user.